See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/332590870

Stewart M. Bracegirdle - DNVGL-RP-F101 Corroded Pipelines

Technical Report · April 2019

CITATIONS READS

0 363

1 author:

Stewart Bracegirdle
Teesside University
2 PUBLICATIONS   0 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

The development of a verification tool to confirm the validity of results of a pipeline corrosion assessment View project

Using the Altman z-score model to test bankruptcy in the Oil Industry View project

All content following this page was uploaded by Stewart Bracegirdle on 23 April 2019.

The user has requested enhancement of the downloaded file.

Contents
Section Page
Abbreviations ........................................................................................................................................ 3
1. Abstract .................................................................................................................................... 4
2. Introduction .............................................................................................................................. 5
3. Aims/Objectives ....................................................................................................................... 7
4. Implementation, Activities and Research ............................................................................. 8
4.1 Oceaneering Pipeline Department corrosion assessment ........................................................ 8
4.2 Overview of corrosion in pipelines and pipeline codes ............................................................. 8
4.2.1 Overview of corrosion ................................................................................................................ 8
4.2.2 Overview of corrosion in pipelines ............................................................................................. 8
4.2.3 Corrosion Orientations ............................................................................................................... 9
4.2.4 Corrosion defect profile - idealisations .................................................................................... 10
4.2.5 Single corrosion defect assessment........................................................................................ 13
4.3 History of assessment methods in corrosion defects and industry codes .............................. 14
4.3.1 History of assessment methods in corrosion defects .............................................................. 14
4.3.2 Overview of ASME B31G ........................................................................................................ 14
4.3.3 Overview of DNVGL-RP-F101 ................................................................................................ 15
4.4 Selecting software for the verification tool ............................................................................... 16
4.5 Learning to use Mathcad ......................................................................................................... 16
4.6 Testing Mathcad calculation sheet .......................................................................................... 17
4.7 Population and sample of the study ........................................................................................ 19
4.7.1 Population ................................................................................................................................ 19
4.7.2 Sampling frame ....................................................................................................................... 19
4.7.3 Sample..................................................................................................................................... 19
4.8 Calculation sheets (Mathcad and Excel) ................................................................................. 20
5. Difficulties encountered ........................................................................................................ 24
6. Results/Outcomes ................................................................................................................. 24
7. Costs ....................................................................................................................................... 25
8. Discussions/Conclusions ..................................................................................................... 25
9. Evaluations ............................................................................................................................ 26
10. References ............................................................................................................................. 27
11. Appendix A – Presentation Slides ....................................................................................... 28
12. Appendix B – Project Log and Gantt Chart ........................................................................ 29
12.1 Project Log .............................................................................................................................. 29
12.2 Gantt Chart .............................................................................................................................. 30
13. Appendix C - Sample results ................................................................................................ 31

Stewart M. Bracegirdle - Mechanical Engineering Project 2 of 42

Abbreviations

ASME The American Society of Mechanical

Engineers
API American Petroleum Institute
DNV GL Det Norske Veritas Germanischer
Lloyd
psi Pound-force per square inch
RP Recommended Practice
FFS Fitness for Service
NDT Non-destructive Testing
NACE National Association of Corrosion
Engineers
SMYS Specified Minimum Yield Strength
UTS Ultimate Tensile Strength
WT Wall Thickness

Stewart M. Bracegirdle - Mechanical Engineering Project 3 of 42

1. Abstract

This report sets out to develop a verification tool to test the validity of corrosion assessment calculation
results of a 20” riser following the guidelines of the DNVGL-RP-F101 - Corroded Pipelines code. The aim is
for the tool to be implemented for use in the Oceaneering Pipelines department to assess fitness for service
of pipelines and risers in the North Sea. At present, there is one Excel calculation sheet for corrosion
assessments; however, there needs to be a method to test the accuracy of its results. Therefore, it is
proposed that a verification tool will be developed in Mathcad to assess single corrosion defects. Once the
tool is finished, testing will be carried out by the Principal Pipeline Engineer to ensure the results are correct.
Ultimately, the aim is to produce a tool that can be used by all the members of the Oceaneering pipelines
team for the assessment of corrosion defects on live pipelines and risers in the North Sea. Knowing that the
corrosion assessment calculations have been verified allows for greater confidence in the team’s results and
therefore suitable engineering recommendations can be made about the fitness for service of the pipeline or
riser under assessment. In conclusion, the Mathcad verification tool has been tested and accurately
assesses corrosion defects. The original results from the Excel calculation sheet on the 20” riser have been
verified and the tool can now be used in the Oceaneering Pipelines team for further validation of corrosion
defect results, thus the aim has been achieved.

Stewart M. Bracegirdle - Mechanical Engineering Project 4 of 42

2. Introduction
I work as an Engineering Assistant within the subsea pipelines team at Oceaneering International
Incorporated in Aberdeen. This role involves providing engineering support to - and document preparation for
- the subsea pipelines team. Oceaneering provides engineering services and products primarily to the
offshore energy industry (Oceaneering, 2018) and in the Aberdeen office, I work within the Asset Integrity
division, specifically in the pipelines team which supports various clients and is involved in the reporting and
assessment of the integrity of their pipelines, risers and subsea structures. Asset Integrity, as a division,
provides services of integrity management of assets to ensure clients have all the data necessary to make
educated decisions about the fitness for service of their equipment. We advise on the frequency of
inspections required to assess the condition of anomalies on pipelines, risers and subsea structures. The
inspections use a mix of NDT methods and visual inspections, chosen based on the nature of the inspection
criteria and location of pipe to be inspected.
I have involvement in writing Inspection and Anomaly Assessment (IAA) reports which provide the client
with: updates on the condition of known anomalies, any new anomalies and specific recommendations to
remediate the anomaly or monitor it in future inspection plans. Primarily, our inspection is focused on the
assessment of adequate internal pressure containment in pipelines; this is done by assessing and
calculating the minimum allowable wall thickness (MAWT). A major consideration in this assessment is
corrosion – both internal and external – which may reduce the wall thickness and therefore the capability of a
pipe to contain internal pressure and fluids.
In order to assess the wall thickness and corrosion, we abide by industry standards. The department follows
‘PD 8010-2:2004, Code of Practice for Pipelines, Part 2, Subsea Pipelines.’ to calculate the minimum
required wall thickness for subsea risers, subsea spools and pipelines. Assessing the minimum required wall
thickness allows engineering judgement to be made about the fitness for service of the equipment under
scrutiny. The results are crucial for the assessment of remaining life and safe working conditions of the
subsea riser, spool or pipeline. To further investigate, and evaluate, the fitness for service of a pipeline it is
prudent to look at other codes in conjunction with PD8010, which is where there is a need for DNVGL-RP-
F101. This code covers the assessment of corroded pipelines and the results allow for recommendations to
be made to continue using a riser, spool or pipeline up to 85% of the wall thickness. This is less conservative
TM
than other assessment standards used in the industry (for example, ASME modified B31G (‘RSTRENG ’))
(Hopkins, P., 2014).
Currently, there is no verification tool in the Oceaneering Pipelines department to validate results of DNVGL-
RP-F101 calculations. At present only a Microsoft Excel sheet is used to do calculations on the fitness for
service of a corroded pipeline, Figure 1 shows an example of an assessment carried out on a 20” riser in the
North Sea. This Excel calculation sheet follows the guidelines of DNVGL-RP-F101 and results are generated
on the fitness for service based on the corrosion defect depth, from a given defect length. However, there is
no other source to confirm that these calculation results are correct. In light of this, the Principal Pipeline
Engineer requested that an additional testing tool be created for the Pipelines department. Not only will this
give greater confidence - and verify - that the assessment results in the Excel sheet are correct but it will also
aid my professional development in terms of using new software and learning the DNVGL corroded pipelines
code.

Stewart M. Bracegirdle - Mechanical Engineering Project 5 of 42

Figure 1: Original 2006 Excel Riser Splashzone Inspection sheet requiring validation

Stewart M. Bracegirdle - Mechanical Engineering Project 6 of 42

3. Aims/Objectives
The aim of the project is to:

• Develop a verification tool in Mathcad to confirm the validity of calculation results of a pipeline
corrosion assessment done on a 20” riser in 2006, following the requirements stipulated in DNVGL-
RP-F101.

In addition to the project aim, several objectives will also be achieved:

• Learning the DNVGL-RP-F101 code – beneficial for professional development in pipeline

engineering.
• Learning Mathcad – software that had not previously been used. It will allow for the development of
additional calculation sheets in future.
• Understanding the DNVGL-RP-F101 code and the codes other societies have created for corrosion
defect assessments (e.g. ASME B31G – Manual for Determining the Remaining Strength of
Corroded Pipelines).

Stewart M. Bracegirdle - Mechanical Engineering Project 7 of 42

4. Implementation, Activities and Research
4.1 Oceaneering Pipeline Department corrosion assessment
In 2006, a calculation sheet was developed in Excel to assess a 20” riser in the splashzone on an offshore
asset in the North Sea - a riser is defined as a “section of pipeline extending from the ocean floor up to an
offshore platform” (NACE, 2002). There was no verification tool to check if the results of allowable defect
length vs. defect depth were correct – instead it was passed from Author to Senior Pipeline Engineer to
Principal Pipeline Engineer to do a thorough check that the results were consistent. This assessment had a
sample of defect lengths from 0mm to 1000mm; the same will be used in the Mathcad calculation sheet
analysis as part of this project in order to clarify that the results are correct.

The DNVGL-RP-F101 Corroded Pipelines code was followed in the original Excel sheet and this can be
used for a variety of pipelines with different material properties (e.g. steel grades A25, B, X42, X46, X52,
X56, X60, X65, X70 and X80). The two digit number after the X shows the minimum yield strength in 1000’s
of psi (PM International Suppliers, 2018). Each pipeline grade has a different SYMS (MPa) and UTS (MPa)
which feed into the equations and determine the defect length vs. defect depth. It must be noted that the
DNVGL-RP-F101 code does not allow assessments for pipeline steel grades above X80. Typically with
North Sea pipelines, the pipeline grades from API 5L are in the range of X52 to X65, thus the majority of
pipelines that are assessed by Oceaneering are well within the constraints of the code (for example, the
pipeline assessed in this project has a steel grade of X60).

4.2 Overview of corrosion in pipelines and pipeline codes

4.2.1 Overview of corrosion
The National Association of Corrosion Engineers (NACE) defines corrosion as “the deterioration of a material
(usually a metal) that results from a chemical or electrochemical reaction with its environment” (NACE,
2018). The implications of corrosion can be significant if not properly monitored and there are substantial
costs associated with the maintenance of corrosion. Worldwide, the cost of corrosion is up to 3.5% of global
GDP annually – this equates to a value of circa $4 trillion (Enegela, P., 2018). In the Oil & Gas production
industry specifically; the yearly cost of corrosion is valued at approximately $1.372 billion (NACE, 2018). The
foremost reason for piping failures is corrosion and the “costs for monitoring, replacing, and maintaining
gathering and transmission pipelines is estimated at $7 billion annually” in the United States (NACE, 2018).
As failure in piping occurs primarily from corrosion, it is crucial that proper maintenance and corrosion
assessment regimes are in place, not only from a financial stand-point but also, and more importantly, for
safety. The implications of a pipeline failure can be catastrophic. Reuters (2013) reported an oil pipeline
failure in China which resulted in 35 people being killed – just one example of the tragic repercussions when
a pipeline fails.
4.2.2 Overview of corrosion in pipelines
Any exposed pipelines are at risk from corrosion (NACE, 2018) and without the proper corrosion
assessments and maintenance in place; all pipelines will ultimately fail and deteriorate in time. The integrity
of a pipeline can be jeopardised and therefore make it not fit for purpose – its purpose being to transport oil,
gas or other hazardous materials. Primarily, Oceaneering’s pipeline department deal with pipelines
submerged in water making these highly susceptible to corrosion. To mitigate the effects of corrosion,
cathodic protection is used (NACE, 2018). According to NACE (2013), Cathodic Protection is “a technique to
reduce the corrosion of a metal surface by making that surface the cathode of an electrochemical cell”. In
order to get the best and longest use from a pipeline, proper corrosion control and monitoring/maintenance
programmes must be in place. In the pipelines department, calculations are done to determine the remaining
wall thickness and therefore the ability for the pipeline to contain internal pressure. Ultimately, the aim is to
increase a pipeline’s fitness for service as long as possible without posing a safety risk.

Stewart M. Bracegirdle - Mechanical Engineering Project 8 of 42

Corrosion is a very complex subject and because it is naturally-occurring, corrosion defects can form in a
variety of different orientations: axial, circumferential, spiral, and random; and these defects are irregularly
shaped. The variety of different orientations makes for difficulties in how to assess defects. Thus,
assumptions/idealisations are made about the dimensions of corrosion defects to allow for easier
assessment – idealisations about defect profiles are discussed below.
Figures 3, 4 and 5 show the different methods of assessment for the irregular shapes of corrosion defects.
Although the actual area of a corrosion defect can be assessed, the corroded pipelines codes tend to use a
parabolic or rectangular defect area. ASME B31G assumes a defect shape which is parabolic, whereas
DNVGL-RP-F101 assumes a rectangular defect area (Hopkins, P. 2014, p.415).

4.2.3 Corrosion Orientations

Figure 2 Corrosion defect orientations (Norman, D. and Argent, C., 2007, cited in Hopkins, P., 2014 p.373)

Stewart M. Bracegirdle - Mechanical Engineering Project 9 of 42

4.2.4 Corrosion defect profile - idealisations
Single defect – rectangular profile, parabolic shape and actual area.

Figure 3 Corrosion defect profile (Hopkins, P., 2014 p.390)

Stewart M. Bracegirdle - Mechanical Engineering Project 10 of 42

Figure 4 Actual area and simplification using rectangular profile (Hopkins, P., 2014 p.391)

Stewart M. Bracegirdle - Mechanical Engineering Project 11 of 42

Figure 5 Simplification using parabolic profile and comparison of profiles (Hopkins, P., 2014 p.392)

Stewart M. Bracegirdle - Mechanical Engineering Project 12 of 42

4.2.5 Single corrosion defect assessment
It is important to ensure that a corrosion defect is isolated and can therefore be treated a single defect.
Therefore, the criteria in the pipeline codes are followed in terms of the minimum separation (circumferential,
ϕ, and axial/longitudinal, s) that ensures two defects are not interacting. If it is found that defects are
interacting, it is no longer possible to use the ‘4.3 Assessment of a single defect’ section of the DNVGL-RP-
F101 code and ‘4.4 Assessment of interacting defects’ is to be used instead. The criteria for assessing
defect interactions are given below and have been used in the calculation sheet (Section 4.8 Figure 9).
A defect can be treated as an isolated defect, and interaction with other defects need not be considered if
either of the following conditions is satisfied (DNVGL, 2017):
1) The circumferential angular spacing between adjacent defects, ϕ:
𝑡𝑛𝑜𝑚
𝜑 > 360√
𝐷𝑜
2) The circumferential angular spacing between adjacent defects, s:

𝑠 > 2.0�𝐷𝑜 × 𝑡𝑛𝑜𝑚

Although, a verification sheet for assessment of interacting defects would be beneficial for use in the
department, there is not adequate time to develop this; however, it does provide an opportunity for further
research/Mathcad calculation sheet development. The intention is to create a Mathcad calculation sheet for
interacting defects in future.

Figure 6 Dimensions required to ensure the defect is single and not interacting (DNVGL, 2017, p. 42)

Stewart M. Bracegirdle - Mechanical Engineering Project 13 of 42

4.3 History of assessment methods in corrosion defects and industry codes
4.3.1 History of assessment methods in corrosion defects
In the 1960s, collaboration between a pipeline company and the Battelle Memorial Institution researched
corrosion defects in pipelines, including investigation into the correlation between defect size and the
threshold internal pressure that would induce defect leakage and/or failure (ASME, 1991). From this
research it would be determined that a methodology/industry standard could be developed to test corrosion
in pipelines that were in use at the time. The results meant an oil company could determine if a pipeline is fit
for service, or if remedial work is required, to ensure safe working operation of the pipeline. Other oil
companies began to take interest in the research and wished to apply the findings to their pipelines to ensure
they were operating safely.
In the early 1970s the American Gas Association (AGA) took on the responsibility for originating ways of
calculating pressure containment capabilities of pipelines with a variety of different corrosion defects. The
objective; to find a correlation between internal pressure containment and corrosion defect size(ASME,
1991).
The B31G code was developed from this research and was based upon the full-scale tests of corroded pipes
that were pressurised until failure. Since there was an abundance of pipelines which had previous been in-
service but were subsequently removed due to corrosion, it was proposed to test these instead of using
pipelines in a lab environment with machined defects. Around 300 tests were conducted on all different types
of defects and from the results, mathematical equations were developed to calculate pressure containment
in corroded pipeline in line with fracture mechanics principles. These principles being: the tougher the
material, the larger the larger a defect can be resisted before failure; and the bigger the defect, the lower the
pressure value at which failure will occur. Although these may appear intuitive, they lead to the deduction of
the strength of a pipeline which contains defects (ASME, 1991).
At the start of the 1970s, pressure-testing of pipelines of varying sizes were done to evaluate how accurate
the mathematical equations were in calculating the strength of pipelines with defects. The diameter range of
these tests was 16” to 30” and wall thicknesses of 0.312” and 0.375”, respectively. A variety of pipes were
used in terms of yield strength from Grade A (yield strength 25,000 psi) to Grade X52 (yield strength 52,000
psi) – in more recent revisions of the B31G and DNVGL-RP-F101 codes, steel grades up to X80 (yield
strength 80,000 psi) can be assessed. The aforementioned mathematical equations have since been
modified to ensure reliable failure results can be found (ASME, 1991). These equations are included in both
ASME B31G and DNVGL-RP-F101 – an overview of these codes follows in sections 4.3.2 and 4.3.3.
4.3.2 Overview of ASME B31G
The ASME B31G ‘Manual for Determining the Remaining Strength of Corroded Pipelines’ code was
developed in 1984 to assess corrosion defects in pipelines with reference to the Battelle corrosion research
from early 1970s (Kiefner, J.F. and Dufy, A. R, 1971). It was deemed that the original B31G code was too
conservative; thus, a modified code was developed in 1989 known as ‘modified B31G’. There were
additional revisions of the code in 2012, 2015 and 2017 whereby levels of assessment were included. For
example, B31G has levels 0, 1, 2 and 3 – level 2 being less conservative than levels 0 and 1. Level 3 is very
detailed and requires finite element stress analysis. Each level is more complex and requires more detailed
measurements of the corrosion plus the use of software to allow for repetitive calculations. The levels give
more accurate calculations and estimations of the failure pressure (Hopkins, P., 2014, pp.408-409)). These
‘old’ methods were, for the most part, authenticated by full-scale assessments on older line pipe steels. The
‘new’ methods (e.g. DNVGL-RP-F101) were authenticated through assessments on modern, high
toughness, line pipe steels (Cosham, A. and Hopkins, P., 2004 p.86).

Stewart M. Bracegirdle - Mechanical Engineering Project 14 of 42

4.3.3 Overview of DNVGL-RP-F101
DNVGL-RP-F101 was first issued in 1999 to provide an alternative method to ASME B31G for assessing
corrosion in pipelines. This approach was less conservative than previous methods and provides guidance
on the assessment of: single or interacting defects, complex-shaped defects and combined loading. This
code allows for assessment of defect depths up to, but not exceeding, 85% of the wall thickness (Hopkins,
P., 2014, p.411). The guidance was made from the amalgamation of a joint industry project between DNV
(now DNVGL) and BG, whereby full scale burst tests and numerical analyses were carried out on pipelines
with machined defects assessing their burst capacity (Hopkins, P., 2014, p.411, Bjørnøy, O. H. et al., 1999).
The combined results of the ‘Line pipe Corrosion Group Sponsored Project’ in the UK and DNV’s own work
fed into, and ultimately led to, the creation of the DNV RP-F101 “Corroded Pipelines” code. Further analysis
of the results was carried out and a database of finite element analysis results was produced (Bjørnøy, O. H.
et al., 1999). From here, equations for the burst capacity were derived and used as the basis for the RP-
F101 code. Further developments were made and an equation for the burst capacity of a single defect was
developed – which is used in the assessment in this project.

Stewart M. Bracegirdle - Mechanical Engineering Project 15 of 42

4.4 Selecting software for the verification tool
There were three potential programmes to use to verify the calculation results in the department’s current
Excel sheet, namely: Mathcad, Mathematica and MatLAB. Although having previous experience using
MatLAB during Undergraduate studies in Physics, it was deemed less applicable for the needs of this
project. It has excellent programming functionality - and the ability to model physics concepts – but this is
beyond the requirements of the verification tool for this project. Mathematica is a powerful tool and is
categorised as a computer algebra system (Wolfram, 2018). Although useful, its ability to produce graphs
from calculations is not as good as Mathcad. Both the aforementioned software would have required gaining
a licence to get access to use as well. Additionally, no one in the department possessed prior experience in
using either MatLAB or Mathematica – thus making the process of checking the verification sheet more
difficult. Oceaneering already had a licence for Mathcad and this is a programme regularly used by
engineers in the team. The Principal Pipeline Engineer has extensive experience of using Mathcad and
could therefore more easily check the verification tool. Mathcad was therefore deemed the most appropriate,
and accessible, tool to develop the calculation sheet.

4.5 Learning to use Mathcad

Having never used Mathcad previously, it was imperative that time was allotted to practice and get used to
the software. In order to get started with the software, the Principal Pipeline Engineer ran through the basic
functions and the method of writing formulae. It was suggested that variables be colour-coded in yellow and
defined components (e.g. the result of the Q value) and calculation solutions in green. This makes for easier
reading and tells anyone using the calculation sheet the variables that can be changed and which are to be
left as they are. The Principal Engineer suggested that this colour-coding is recognised in the industry and
following this method would be beneficial for any other calculation sheets that may be created in future.
Additionally, since the colour-coding is used by others in the industry it will also prove useful for any external
party reviewing the sheet(s).
It took a lot of trial and error, and reference to the ‘Quick Sheets’ help in Mathcad, to get used to using the
software and to navigate the toolbars, using graphical notation and locate units. Using Mathcad was done
during working hours so time spent on it was limited each day; however, each time using it a greater
familiarisation with its functions and usability was developed. What was clear was the power of the software
and its advantages; specifically the professional looks of the calculation sheet, and equations, makes for
easier reading. There are a wealth of units, graphs and pre-defined variables in Mathcad – these all prove
useful when developing a calculation sheet. For example, one can change the units from metres to
millimetres by entering mm at the end of a solution; the programme will automatically recalculate the solution
and the decimal places to suit – saving time and unnecessary additional unit conversion calculations – which
would be required in Excel.
Within the DNVGL-RP-F101 Corroded Pipelines code, there are appendices after the theory where
examples are given to aid with the understanding of the code. To test familiarity with - and ability to use –
Mathcad, it was decided that the example in ‘Appendix B Examples for Part B’ should be replicated. This
example - ‘Example 7’ in the code – follows the process for a single defect assessment of a pipeline of steel
grade X65 (a steel grade commonly used for subsea pipelines). The comparison of the code and the
calculation sheet in Mathcad is shown in Figures 7 and 8. As it can be seen, the solutions are the same and
thus prove that there is enough understanding of both the code and Mathcad to proceed to develop the
calculation sheet. The riser splash zone inspection requires more advanced calculations and a deeper
understanding of the RP-F101 code; however, the aim of the project is to advance the department’s
capabilities to validate results and for professional development so learning more technical calculations is
vital and necessary.

Stewart M. Bracegirdle - Mechanical Engineering Project 16 of 42

4.6 Testing Mathcad calculation sheet

Figure 7: Mathcad test calculation sheet following Appendix B Example 7 from DNVGL-RP-F101 Corroded
Pipelines p.79. Result for safe working pressure of corroded pipe agrees with result from code, shown in
Figure 8 below.

Stewart M. Bracegirdle - Mechanical Engineering Project 17 of 42

Figure 8: Appendix B Example 7 from DNVGL-RP-F101 Corroded Pipelines p.79 (DNVGL, 2017).

This test sheet was prepared to confirm working knowledge and familiarity of Mathcad. As the resulting safe
working pressure value from the worked example in Figure 7 from the code and the Mathcad sheet in Figure
2
8 are the same (i.e. ~10.28 N/mm ) it can be confirmed that there is sufficient understanding of the software
to proceed with developing the validation tool.

Stewart M. Bracegirdle - Mechanical Engineering Project 18 of 42

4.7 Population and sample of the study
4.7.1 Population
The population of a study is defined in slightly different ways by academics. Bryman and Bell (2011) describe
it as “the universe of unity from which a sample is to be selected”; Lewis, Saunders and Thornhill (2012)
describe it as “the complete set of cases or group members” (Bryman and Bell, 2011; Lewis, Saunders and
Thornhill, 2012). Although these definitions vary, the general idea of a population is the complete group of
the test subject. It would not be practical, given the limited time for the project to analyse the complete
number of sample defects lengths/depths in the original Excel sheet study. Thus, a sample is selected from
the population of 1000 defect lengths - a sample is a smaller “sub-group or part of a larger population”
(Lewis, Saunders and Thornhill, 2012). This alleviates the impracticability of testing a population of hundreds
of defect lengths in the limited time available. An appropriate sample was selected to represent the
population. This was prepared under the associate method of probability (or representative) sampling,
whereby implications need to be made from the selected sample, about the population. The sample is
prepared in order to address, and confirm, the validity of workability of the Mathcad sheet. This will ultimately
lead to confirming the objectives set by the researcher are met.

4.7.2 Sampling frame

To allow a sample to be selected, the sampling frame was established. The sampling frame is a “complete
list of all the cases in the population from which your sample will be drawn” (Lewis, Saunders and Thornhill,
2012). To allow generalities to be drawn from the sample, about the population as a whole, the defect
lengths needed to be taken at random to give an overall confirmation that results are correct and the
Mathcad sheet is accurately calculating the defect depth from its given length. Considering the total number
of defect lengths in the original Excel sheet gives a population of 1000 lengths.
4.7.3 Sample
The most suitable method to test the results in the Excel sheet with the Mathcad calculation sheet is to take
a sample length, L, between the ranges of 0mm to 1000mm. The proposed sample follows systematic
sampling to assess n = 20 different lengths. Thus, with a population of x = 1000 lengths, the length will be
taken at 50 length intervals (i.e. x/n = 50). The sample will be as follows (49mm, 99mm, 149mm,
199mm…998mm). There are no lengths of 999mm or 1000mm recorded in the Excel sheet, hence 998mm is
the last assessment length to check. In addition, lengths were selected at random (results are given for
lengths 467mm and 652mm). Both randomly-chosen lengths gave the same result as found in the original
Excel sheet. Therefore, it can be said, with confidence, that the Mathcad sheet is indeed correctly calculating
the depth of defect from its length. Sample defect lengths were entered into the Mathcad sheet to see if the
defect depth df101 (column AO) in the Excel sheet is the same (see Appendix C for results).

Stewart M. Bracegirdle - Mechanical Engineering Project 19 of 42

4.8 Calculation sheets (Mathcad and Excel)

Stewart M. Bracegirdle - Mechanical Engineering Project 20 of 42

Figure 9: Mathcad calculation sheet testing the figures from the original Excel Riser Splashzone Inspection
sheet with details of where each equation was found.

Stewart M. Bracegirdle - Mechanical Engineering Project 21 of 42

Figure 10: Original 2006 Excel Riser Splashzone Inspection sheet requiring validation

Stewart M. Bracegirdle - Mechanical Engineering Project 22 of 42

Figure 11: Graph of allowable defect length vs. defect depth which shows the same curvature as the graph
from the original Excel sheet.

The graph was generated from all the defects from 0mm to 1000mm (L:=0,1mm..1000mm) from Figure 8.
Using this formula in Mathcad generates a list of numbers of this range. Excel can more easily generate a list
of figures as an equation and its variables can be dragged down cells. For Mathcad, it took time to figure out
how to get a similar list of figures but once generated, the list could be used for the range on the x-axis for
defect lengths. To get the corresponding variables for the y-axis formulae were entered into Mathcad. These
can be seen in Figure 9 (Def(L):= Find (d) and Cor(L):= if(Def(L) ≥ 0.85*tnom, 0.85*tnom,Def (L)). This created
corresponding figures for the depths corresponding to the range of lengths from 0mm to 1000mm. The graph
of these is then generated and the curve of the graph is seen to match that in the original Excel sheet. A
horizontal line for corrosion allowance and a vertical line for the assumed corrosion length were added to the
graph for context and to be consistent with the original Excel sheet graph.

Stewart M. Bracegirdle - Mechanical Engineering Project 23 of 42

5. Difficulties encountered
Not having used Mathcad previously; nor having an understanding of the code proved to be the most
challenging and time-consuming aspects of the project. Being new to the pipeline engineering sector in the
oil and gas industry meant there was a significant number of principles and codes to learn. To overcome the
challenge of using Mathcad, the Principal Engineer kindly spent time going through functions of Mathcad and
also advised that the ‘Quick Sheets’ are a great resource included in the software. The Quick Sheets provide
worked examples on a range of common questions users of Mathcad have. This project proved to be
mutually beneficial in terms of furthering understanding of pipeline codes and also providing the pipelines
department with a calculation sheet to verify results of a corrosion assessment.

6. Results/Outcomes
Tests of the Mathcad sheet were conducted by the Principal Pipeline Engineer and the conclusion was
drawn that the sheet worked and can correctly validate the results of earlier corrosion assessments in Excel.
The aim is to be able to use the Mathcad sheet in conjunction with the current Excel sheet to ensure all the
results are accurate.
Ultimately, to test the true usability and accuracy of the calculation sheet, it will need to be used for real-time
pipeline assessment. It is critical that the tool works correctly on the assessment of the 20” Riser Splashzone
test and is accurate to ensure the safety and integrity of any assessed pipelines in future. Since the code
allows for up to 15% remaining wall thickness, there are significant safety implications if any erroneous
results are obtained and used in fitness for service calculations.
The confirmation of the sheet being accurate by the Principal Pipeline Engineer is adequate for use in the
department but only once this is being used for real-time assessment will it confirms its true usefulness.
The Mathcad sheet has the capability to be used on pipelines and risers of varying diameters, steel grades,
internal pressures and safety factors. Thus, the tool is versatile and covers the majority of corrosion
assessments being carried out on a clients’ subsea pipeline infrastructure.

Stewart M. Bracegirdle - Mechanical Engineering Project 24 of 42

7. Costs
The project itself did not have any associated costs apart from the time spent during working hours to use
Mathcad and study the code. It was deemed appropriate to use working hours for this project as the tool will
be used in the department. The financial benefits of the project could be significant, if a pipeline or riser’s
working life can be extended, there would be less need for remedial work on said components. The
extension in working life means there is: less inspection work offshore, fewer man-hours for inspections and
a longer period of time for flow of fluids through the component.
No immediate financial benefit will be gained from the project. However; with validated results, and therefore
confidence in the corrosion assessments, engineering recommendations can be made that a pipeline is
indeed fit for service and can safely contain internal pressure. The financial benefit of this would mean that a
pipeline previously considered not fit for service at e.g. 80% of wall thickness, could now be deemed fit for
service up to 85% of the wall thickness – increasing its usable life.

8. Discussions/Conclusions
The Principal Pipeline Engineer has confirmed the sheet as working by checking it produces accurate and
consistent solutions. Within the department, reports will go through a process of being checked prior being
issued to the client. Thus, it was prudent to follow a similar process to get the method and outcomes of sheet
tested and confirmed to be accurate and in-line with what is needed in the department.

I have learned the single-defect assessment sections of the DNVGL-RP-F101 code and this knowledge can
be used to run corrosion assessments and also to aid fellow colleagues by sharing my experience using the
calculation sheet. I learned to use Mathcad and intend to create further calculation sheets to verify the
results of minimum wall thickness calculations (currently done in Excel) and for interacting corrosion defects
following the DNVGL-RP-F101 code.
I addition, I have gained knowledge of other industry codes alongside DNVGL-RP-F101. Namely, the ASME
B31G – Manual for Determining the Remaining Strength of Corroded Pipelines code. These codes are vast
in scope and length so I will continue to delve deeper to ensure my understanding is sound. Sharing what I
have learned with my colleagues will be a highly beneficial way to test my knowledge and ability to convey
the content of the codes.

Stewart M. Bracegirdle - Mechanical Engineering Project 25 of 42

9. Evaluations

It took a significant amount of time to choose the project topic. Initially, the topic proposed was to write a
report on the failure of a subsea isolation valve (SSIV) and an actuator changeout; however, it was deemed
there would not be adequate assessment of engineering ability. Instead, this topic was more about writing up
the findings of a failure report from an offshore platform. The Principal Pipeline Engineer then suggested that
the department needed a corrosion assessment verification sheet and it would involve enough application of
engineering concepts and the use of software to fulfil the expectations of the project.

Once the project topic was decided, time frames were made and entered into a Gantt chart (Appendix B
section 12.2). This gave the basis and guidelines of how to ensure the project was completed in an
appropriate time. Having milestones kept motivation high and increased productivity.

Although the initial time to get familiar with Mathcad was longer than planned, time needed to be spent on
this aspect of the project, otherwise, there would not have been a chance that the calculation sheet would be
correctly developed. Working on Mathcad in conjunction with reading the DNVGL-RP-F101 corroded
pipelines code helped develop a deeper understanding of the code.

It would have been beneficial to have tried Mathcad prior to starting the project as it would have allowed
more time to develop the calculation sheet and perhaps enough time to develop a sheet to test for interacting
defects – something which would be recommended to do in future when time permits.

Stewart M. Bracegirdle - Mechanical Engineering Project 26 of 42

10. References
ASME (1991). ASME B31G-1991: Manual for Determining the Remaining Strength of Corroded Pipelines.
Revision of ANSI/ASME B31G-1984. The American Society of Mechanical Engineers, U.S.A
Bjørnøy, O. H., et al. (1999). Background to DNV RP-F101 “Corroded Pipelines”. OMAE99/PIPE-5031. Det
Norske Veritas AS, Newfoundland, Canada.
rd
Boboian, R. (2002). NACE Corrosion Engineer’s Reference Book. 3 ed: Houston: NACE INTERNATIONAL.
rd
Bryman, A. and Bell, E. 2011. Business research methods. 3 ed: Oxford: Oxford University Press.
Cosham, A. and Hopkins, P. (2004) Penspen Ltd. Pipeline Defect Assessment Manual (PDAM) A Report to
the PDAM Joint Industry Project. Newcastle upon Tyne.
DNVGL (2017) DNVGL-RP-F101 Corroded Pipelines. Available at:
https://www.dnvgl.com/oilgas/download/dnvgl-rp-f101-corroded-pipelines.html (Accessed: 15 November
2018)
Enegela, P., (2018). Corrosion Awareness Day. Institute of Corrosion. Available at:
https://sites.google.com/site/icorrabz/resource-center (Accessed 6 December 2018).
Hopkins, P. (2014) Penspen Ltd. Defect Assessment in Pipelines Course (PDAC) Part 1. Newcastle upon
Tyne.
Kiefner, J. F. and Duffy, A. R., (1971). Summary of research to determine the strength of corroded areas in
pipe. Columbus Laboratories, Battelle Memorial Institute.
th
Lewis, P., Saunders, M., and Thornhill, A. 2012. Research methods for business students. 6 ed: Harlow:
Pearson.
NACE (2013) Standard Practice: Control of External Corrosion on Underground or Submerged Metallic
Piping Systems (SP0169-2013). Available at:
https://www.nace.org/uploadedFiles/Corrosion_Central/Industries/SP016913%20for%20State%20of%20New
%20York.pdf (Accessed: 15 November 2018).
NACE (2018) Corrosion 101. Corrosion – A Natural but Controllable Process. Available at:
https://www.nace.org/corrosion-101/ (Accessed: 20 November 2018)
NACE (2018) Oil & Gas Production. Available at: https://www.nace.org/Corrosion-Central/Industries/oil---
Gas-Production/ (Accessed: 20 November 2018).
NACE (2018) Pipelines & Underground Systems. Available at: https://www.nace.org/Pipelines-Tanks-
Underground-Systems/ (Accessed: 20 November 2018).
NACE (2018) Cathodic Protection. Available at:
https://www.nace.org/uploadedFiles/Corrosion_Central/Pipeline%20Corrosion.pdf (Accessed: 07/12/18])
Norman, D. and Argent, C. (2007). Corrosion at spiral weld due to tenting of tape. Pipeline Coatings,
External Corrosion and Direct Assessment. NACE Corrosion Conference, 2007.
Oceaneering International Incorporated (2018) Available at: https://www.oceaneering.com (Accessed: 12
November 2018)
PM International Suppliers, LLC. (2018) API 5L X Grades. Available at: https://www.api5lx.com/api5lx-grades
(Accessed 6 December 2018).
Reuters (2013) Sinopec oil pipeline blast kills 35 in eastern China. Available at:
https://www.reuters.com/article/us-sinopec-blast/sinopec-oil-pipeline-blast-kills-35-in-eastern-china-
idUSBRE9AL08E20131122 (Accessed: 20 November 2018).
Wolfram (2018) Wolfram Mathematica. Available at: www.wolfram.com/mathematica (Accessed: 15
November 2018)

Stewart M. Bracegirdle - Mechanical Engineering Project 27 of 42

11. Appendix A – Presentation Slides

Stewart M. Bracegirdle - Mechanical Engineering Project 28 of 42

12. Appendix B – Project Log and Gantt Chart
12.1 Project Log

Date Activity
30/10/2018 Reading and learning the DNVGL-RP-F101
code
01/11/2018 Discussion of project topic with Principal
Engineer
03/11/2018 Consideration of which software to use to
test results
04/11/2018 Introduction to Mathcad and familiarisation
with software
11/11/2018 Working through examples from RP-F101
16/11/2018 Development of Mathcad sheet to do
worked examples from the code
20/11/2018 Developing calculation sheet based on
original Excel riser inspection
28/11/2018 Finished Mathcad sheet, began testing it
28/11/2018 Contact with module tutor regarding project
suitability
30/11/2018 Confirmation from tutor that project is okay
and advice given on the expectation of the
results
01/12/2018 Research for information for inclusion in
report
05/12/2018 Further testing of calculation sheet
18/12/2018 Start of writing the project report
31/01/2019 Presentation of project (video recording)

Table 1: Project Log

Stewart M. Bracegirdle - Mechanical Engineering Project 29 of 42

12.2 Gantt Chart

Stewart M. Bracegirdle - Mechanical Engineering Project 30 of 42

13. Appendix C - Sample results

Figure 12: Sample 49 of 1000 on original Excel sheet

Stewart M. Bracegirdle - Mechanical Engineering Project 31 of 42

Figure 13: Sample 49 of 1000 comparison result on Mathcad sheet

Stewart M. Bracegirdle - Mechanical Engineering Project 32 of 42

Figure 14: Sample 99 of 1000 on original Excel sheet.

Stewart M. Bracegirdle - Mechanical Engineering Project 33 of 42

Figure 15: Sample 99 of 1000 comparison result on Mathcad sheet

Stewart M. Bracegirdle - Mechanical Engineering Project 34 of 42

Figure 16: Sample 199 of 1000 on original Excel sheet.

Stewart M. Bracegirdle - Mechanical Engineering Project 35 of 42

Figure 17: Sample 199 of 1000 comparison result on Mathcad sheet

Stewart M. Bracegirdle - Mechanical Engineering Project 36 of 42

Figure 18: Sample 998 of 1000 on original Excel sheet.

Stewart M. Bracegirdle - Mechanical Engineering Project 37 of 42

Figure 19: Sample 998 of 1000 comparison result on Mathcad sheet

Stewart M. Bracegirdle - Mechanical Engineering Project 38 of 42

Figure 20: Randomly selected sample 467 of 1000 on original Excel sheet.

Stewart M. Bracegirdle - Mechanical Engineering Project 39 of 42

Figure 21: Randomly selected sample 467 of 1000 comparison result on Mathcad sheet

Stewart M. Bracegirdle - Mechanical Engineering Project 40 of 42

Figure 22: Randomly selected sample 652 of 1000 on original Excel sheet.

Stewart M. Bracegirdle - Mechanical Engineering Project 41 of 42

 Figure 23: Randomly selected sample 652 of 1000 comparison result on Mathcad sheet

Stewart M. Bracegirdle - Mechanical Engineering Project 42 of 42

View publication stats